@furifont/dictionary 0.1.2 → 0.2.0

package/dist/index.d.mts

@@ -1,5 +1,5 @@
 import { Dictionary } from "@furifont/core";
 //#region src/index.d.ts
-/** Loads the compiled dictionary: JMdict and JMnedict place names, aligned by JmdictFurigana, with KANJIDIC2's on'yomi and JLPT levels. */
+/** Loads the compiled dictionary: JMdict and JMnedict place names, aligned by JmdictFurigana, with KANJIDIC2's on'yomi, JLPT levels and uses in Tatoeba's sentences. */
 export declare function loadDictionary(): Promise<Dictionary>;
 //#endregion

package/dist/index.mjs

@@ -1,18 +1,20 @@
 import * as v from "valibot";
-//#region src/defaults.ts
+//#region src/overrides.ts
 /**
-* Default readings the compiled data ranks wrong, in inline notation. JMdict's
+* Readings that override the compiled data where it ranks them wrong, in inline notation. JMdict's
 * frequency tags belong to a written form, so across entries they cannot tell
 * two readings apart; and a form that drops its okurigana usually meets it in
 * running text, where the font cannot see it. Users' pins override these.
 */
-const DEFAULT_READINGS = [
+const OVERRIDE_READINGS = [
 	"出[で]来[き]",
 	"何[なん]人[にん]",
 	"何[なん]時[じ]",
 	"三[さん]十[じゅう]日[にち]",
 	"外[そと]付[づ]",
-	"支[し]払[はら]"
+	"支[し]払[はら]",
+	"字[じ]",
+	"明[みん]朝[ちょう]"
 ];
 /**
 * Kanji left bare when alone in a run, though one reading dominates the
@@ -50,6 +52,11 @@ const JlptSchema = v.array(v.tuple([
 		5
 	])
 ]));
+const TatoebaSchema = v.array(v.tuple([
+	v.string(),
+	v.string(),
+	v.pipe(v.number(), v.integer(), v.minValue(1))
+]));
 /**
 * A data file as a JSON module, which Node loads and bundlers split into a
 * lazy chunk. The path is a template so that TypeScript leaves the 11 MB file
@@ -59,18 +66,20 @@ async function readData(name, schema) {
 	const { default: data } = await import(`../data/${name}.json`, { with: { type: "json" } });
 	return v.parse(schema, data);
 }
-/** Loads the compiled dictionary: JMdict and JMnedict place names, aligned by JmdictFurigana, with KANJIDIC2's on'yomi and JLPT levels. */
+/** Loads the compiled dictionary: JMdict and JMnedict place names, aligned by JmdictFurigana, with KANJIDIC2's on'yomi, JLPT levels and uses in Tatoeba's sentences. */
 async function loadDictionary() {
-	const [data, places, jlpt] = await Promise.all([
+	const [data, places, jlpt, tatoeba] = await Promise.all([
 		readData("dictionary", DataSchema),
 		readData("places", PlacesSchema),
-		readData("jlpt", JlptSchema)
+		readData("jlpt", JlptSchema),
+		readData("tatoeba", TatoebaSchema)
 	]);
 	const levels = new Map(jlpt.map(([form, furigana, level]) => [`${form}|${furigana}`, level]));
+	const uses = new Map(tatoeba.map(([form, furigana, n]) => [`${form}|${furigana}`, n]));
 	return {
 		sources: data.sources,
 		onyomi: data.onyomi,
-		defaults: DEFAULT_READINGS,
+		defaults: OVERRIDE_READINGS,
 		bare: BARE_FORMS,
 		places: places.map(([form, furigana]) => ({
 			form,
@@ -78,6 +87,8 @@ async function loadDictionary() {
 		})),
 		entries: data.entries.map(([form, furigana, nf, priority, order, conjugates]) => {
 			const level = levels.get(`${form}|${furigana}`);
+			const n = uses.get(`${form}|${furigana}`);
+			uses.delete(`${form}|${furigana}`);
 			return {
 				form,
 				furigana,
@@ -85,7 +96,8 @@ async function loadDictionary() {
 				priority,
 				order,
 				...conjugates && { conjugatesReading: true },
-				...level && { jlpt: level }
+				...level && { jlpt: level },
+				...n && { tatoeba: n }
 			};
 		})
 	};

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@furifont/dictionary",
   "type": "module",
-  "version": "0.1.2",
+  "version": "0.2.0",
   "description": "JMdict words aligned with per-kanji readings, compiled for Furifont",
   "author": "Anthony Fu <anthonyfu117@hotmail.com>",
   "license": "CC-BY-SA-4.0",
@@ -25,11 +25,15 @@
     "dist"
   ],
   "dependencies": {
-    "@furifont/core": "0.1.2",
+    "@furifont/core": "0.2.0",
     "valibot": "^1.5.0"
   },
+  "devDependencies": {
+    "@furifont/compiler": "0.2.0"
+  },
   "scripts": {
     "build": "tsdown",
-    "update": "node scripts/update.ts"
+    "update": "node scripts/update.ts",
+    "audit": "node scripts/audit.ts"
   }
 }